Ethanol is the most commonly consumed drug in the world. Atrial fibrillation (AF) is the most common arrhythmia: it currently affects several million Americans and is responsible for substantial morbidity and mortality with an estimated annual health care cost greater than 6.5 billion dollars. Small observational studies have suggested that ethanol can trigger acute episodes of AF. Large, prospective, epidemiological studies suggest that ethanol use may result in new-onset AF. However, the mechanisms underlying the potential association between either acute or chronic ethanol exposure and AF remain unknown. Investigating these causal mechanisms is important for two reasons: first, evidence of a causal association between ethanol and AF would be pertinent to the > 100 million Americans that consume ethanol. Given the common notion that ethanol is heart healthy, demonstrating that ethanol has electrophysiological and/ or structural cardiac effects that promote AF would be particularly important. Second, understanding how an external source can acutely trigger or chronically lead to the development of AF would reveal common mechanisms underlying AF in general. There is currently no experimental model wherein human AF can be reliably triggered and no known method to prevent AF. This research could therefore open up a new field of experimental human research in AF that ultimately leads to novel therapies or clinical strategies targeting the processes responsible for AF. In Aim 1, we will perform a randomized trial of intravenous ethanol versus placebo in paroxysmal AF patients undergoing invasive ablation procedures to determine the acute electrophysiological effects of ethanol on in vivo myocardium. We will compare pre and post-infusion premature atrial contraction counts, refractory periods, conduction velocities, activation-recovery intervals, restitution properties, ad susceptibilities to induced AF. In order to assure a consistent blood ethanol concentration between patients and within-patients during the experiment, we will employ an established pharmacokinetic model to titrate and then clamp the ethanol infusion to maintain a steady blood ethanol concentration determined by serial breath tests. The mechanisms will be furthered elucidated in an animal model utilizing optical mapping. In order to assess the chronic effects of ethanol, Aim 2 will involve a secondary analysis of serial ethanol assessments and echocardiograms in the Framingham Heart Study to determine if ethanol-induced left atrial enlargement is responsible for incident AF. An animal model of chronic ethanol consumption will be used to further elucidate underlying mechanisms. In Aim 3, we will examine the real-time association between oral ethanol intake and AF episodes in paroxysmal AF patients wearing an automatically recording electrocardiographic monitor paired with a transdermal ethanol sensor for a four week period. The strength of the association between acute ethanol intake and AF episodes as well as the nature of the heart rhythm prior to ethanol-associated episodes (including heart rate, heart rate variability, and premature atrial contraction counts) will be determined.